But on the other hand, racism, conscious and otherwise, is still a significant enough force throughout the country that applying strictly race-blind criteria in order to assessing how disadvantaged candidates are is liable to return bad estimates in many cases.
Even if racism is an issue, using indicators of disadvantage such as income/wealth in addition to race (as probably the best proxy of concious or unconcious racism) would almost certainly yield better results than using race alone.
Note that if you do care about the race balance alone, then affirmative action is the best approach, as that will lead to admitting marginally weaker students in less represented races (while only rejecting marginally qualifed applicants in overrepresented groups), as needed to reach the desired share.
This is true, but it seems to me that buybuydandavis would prefer a system that attempts to isolate causes of privilege independent of race, and my point is that this is likely to be a lot harder than one which isn’t race-blind.
Even if racism is an issue, using indicators of disadvantage such as income/wealth in addition to race (as probably the best proxy of concious or unconcious racism) would almost certainly yield better results than using race alone.
Note that if you do care about the race balance alone, then affirmative action is the best approach, as that will lead to admitting marginally weaker students in less represented races (while only rejecting marginally qualifed applicants in overrepresented groups), as needed to reach the desired share.
I don’t see why this is a reasonable thing to care about.
This is true, but it seems to me that buybuydandavis would prefer a system that attempts to isolate causes of privilege independent of race, and my point is that this is likely to be a lot harder than one which isn’t race-blind.